Segregating the compute cloud

As your cloud infrastructure grows in size, you need to devise strategies to maintain low latency in the API services and redundancy of your service. To cope with unplanned downtime due to natural forces or based on the hypervisor capability itself, the operator must plan for service continuity.

OpenStack Nova provides several concepts that help you segregate the cloud resources. Each segregation strategy brings in its own advantages and shortcomings. We must have a discussion to understand the tools available with an OpenStack cloud operator to manage the need for scale and availability of compute servers.

Availability zones

The concept of Availability Zones (AZ) in Nova is to group together compute nodes based on fault domains: for example, all compute nodes hosted on a rack in the lab. All the nodes connect to the same Top-of-Rack (ToR) switch or are fed from the same Power Distribution Unit (PDU), and can form a fault domain as they depend on a single infrastructure...